I was just reading Jeremy Zawodny's blog entry about "secure IT jobs" (nice poke, btw), and suddenly flashed on my old theory of expanding and contracting IT organizations.
I postulated this after having seen this happen across a number of companies in the valley. Here's how you can tell:
1. Does the engineering organization have its own IT group that *doesn't* report into corporate IT?
2. Do you have a system where groups that are politically powerful enough get their own service organizations, and everyone else has to deal with corporate?
Sound familiar? This is how I look at the chain of events, based on way too many empirical observations:
1. IT organization is running seriously behind technology/customer service curve. Engineering, deals with it by splitting off its own organization
2. Engineering organization manages to get funding that engineering usually would have paid into corporate. Since they can spend all of this on themselves, they go out and get lots of neat toys. They hire in sys admins, tools geeks, the whole works.
3. IT finds themselves further behind the curve, and worse, other groups who can figure out a way to get their own org or to convince engineering to do this do so.
4. Corporate initiative is started to "bring rest of campus up to engineering standards". Usually, this gets about 60% of way there in 200% of the alloted time.
5. The really good corporate IT people have all found their ways into engineering IT positions.
6. Corporate IT makes a case for "global functions should be handled by corporate IT" and demands that the engineering group be folded in to corporate.
7. Groups that don't belong to politically powerful groups do get folded into corporate IT. A small core of sys and net admins are left behind in engineering to deal with "departmental issues". Corporate IT promises all services will be kept up to standards (with a cash infusion of what engineering was paying for their own services)
8. Six months later the folded in engineers are laid off. Engineering is told that if they want to continue to get the same level of services, they will have to pay for contractors, and those contractors will be assigned to them, and will do only tasks for engineering.
9. They aren't--and their cycles keep bleeding off to corporate IT boondoggles.
10. Services call behind the curve, and engineering looks around and sees that their small group of support staff are doing some pretty wild things. Plans are made to expand their functions.....
And so on. I'm just using engineering as a placeholder--I've seen this done with a sales support org as well.
And I suspect that "move jobs offshore" ought to be inserted somewhere into step 8, but it's really it's own contraction/expansion cycle. More on that later....
Posted by lsefton at September 17, 2003 12:46 PMright on.
I've seen a similiar cycle with server consolidation/expansion as well as pretty close to what you're describing in at least two companies.
Posted by: Joe at September 18, 2003 05:41 PM